TR2023000880A1 - A SINGLE CATALOGUE PRODUCT UPDATE SYSTEM AND A RELATED METHOD - Google Patents

Abstract

Our invention specifically; To improve the content quality of products in single catalogs used on e-commerce sites, the product quality system (10) calculates the product content quality score using machine learning method (101), shows previous product content quality scores to vendors (20) or integrators (30) (102), vendors (20) or integrators (30) transmit the product content update request to the management system (70) via user interface (K), document (D) or API (A) (103), enriches the update content by the web search robot (40) (104), processes the incoming update content in the validation system (50) (105), calculates the vendor content score for authorization control purposes (106), the management system (70) decides on update authorization by considering the vendor content score (107), the management system (70) decides on rejecting the update request or directing it to operation (80) (108), approves the update request or directs it to operation for manual control (109), the management system (70) or This relates to a single catalog product update system (1) which improves the product content quality in single catalogs following the processes of sending the update data to the single product catalog (90) for the requests approved by the operation (80) team and obtaining the result (201), recalculating the product content quality score of the product quality system (10) (202) and showing/communicating the request result to the vendor/integrator (203).

Description

DESCRIPTION OF A SINGLE CATALOG PRODUCT UPDATE SYSTEM AND A RELATED METHOD Technical Field Our invention relates to a single catalog product update system and a related method aimed at improving the content quality of products in single catalogs used on e-commerce sites. More specifically, it relates to a system that enables the updating of product information by the same or another seller in order to maximize the sales of a product whose basic information has been entered into the catalog by a seller, and a method that includes algorithmic processes among the components that make up this update system. State of the Art Commerce is defined as the exchange of goods for other economically valuable objects (money, etc.) during the process from the production process to the consumption process, which arose to meet the needs of people. Trade, whose foundations were laid in ancient civilizations through barter, is now conducted in many different ways. Especially with the development of technology, the concept of e-commerce has emerged. E-commerce, an abbreviation of electronic commerce, is defined as trade being conducted in electronic environments. Lists or similar structures prepared in a specific order are called catalogs. In other words, it is the process of bringing together objects that serve the same purpose in an organized manner. This process can be done in different ways, the most well-known being alphabetical sorting. In a commercial sense, the word catalog is very important. Sellers use catalogs to sell their products in a specific order, and customers use them to easily find the products they want to buy. The fact that e-commerce is conducted online creates a need for online catalogs. E-commerce companies, especially those operating as marketplaces, prefer this catalog structure. These catalogs contain products and information about those products (product name, product description, product specifications, etc.). This information is entered into the system by the sellers. Another type of online catalog, the single catalog system, is based on the concept of having only one copy of each product in the catalog and different sellers selling the product at different prices. The information content of the products (product name, product description, product specifications, etc.) is one of the most important factors contributing to the sale of these products. This product information content cannot be updated in some catalog systems within the known state of the art. This situation poses a risk for situations where multiple sellers sell the same product. Because the entered product information may be incorrect/inaccurate, or the entered product information may need to be updated from time to time. In conclusion, the existence of the negative situations mentioned above reveals the necessity of our invention. Purpose and Brief Description of the Invention The purpose of the invention is to provide a single catalog product update system aimed at improving the content quality of products in single catalogs used on e-commerce sites. Another purpose of the invention is to provide a method for improving the content quality of products in single catalogs used on e-commerce sites. The invention is a single catalog product update system for use in single catalogs within e-commerce sites, and it aims to improve the content quality of products in single catalogs and consequently maximize product sales by considering: product name; number and quality of images; presence or absence of video; product description length; whether the product description contains images, videos, tables, and lists; The system takes data input indicating whether or not to enter product type-based attribute values (color, size, etc.), then inputs data on how many sales were made based on this data, calculates a product content quality score once for each category in a single product catalog to determine which data contributed to product sales, based on the two different data points mentioned in the previous two sentences, transmits the calculated score as output to the management system via a message unit, and for products whose update request is approved and updated in a single product catalog, a new product quality system, using the message unit from the aforementioned management system, conducts online research, including competitors, to add to the updated product content information of a vendor or integrator, enriches the product content based on the research results, and transmits the enriched product content to the management system via a message unit. A web search engine is a validation system that checks (forbidden word checks, inappropriate image checks, etc.) the updated product content – which is received from the aforementioned management system using a message unit and then enriched by a vendor or integrator – before adding it to a single product catalog. The system then transmits the results of this check to a management system via a message unit. The system also includes the data entered by the vendor for the relevant product; the number of sales of the relevant product by the vendor; and customer ratings given to the vendor. The invention includes a vendor product scoring system that takes the vendor's average shipping time data as input, calculates a vendor content score to determine whether the updater (vendor or integrator) has the necessary authorization based on the received data inputs, and transmits this calculated score as output to the management system via a message unit. The management system then decides whether to: - forward the update request made by the vendor or integrator to an operations team via a service for manual verification in data format, or - if positive, transmit the updated product information to a single product catalog via a message unit. This invention is a single catalog product update method for use in single catalogs within e-commerce sites, aiming to improve the product content quality in single catalogs and consequently maximize product sales. The product quality system includes: - product name; number and quality of images; The system should take into account the following: the presence of a video; the length of the product description; the inclusion of visuals, videos, tables, and lists in the product description; the entry of product type-based attribute values (color, size, etc.); and, by taking the data and subsequently the number of sales based on that data as input, calculate the product content quality score once for each category in a single product catalog to determine which data contributed to product sales. This product quality system should be applied to products already in the aforementioned single product catalog, considering different product types, product name, number of visuals, description content, etc. The system displays a pre-determined product content quality score, based on fixed scores for each area, to the vendor or integrator via a service in data format. Vendors can update a single product via a user interface in data format, or perform bulk product updates via a digital spreadsheet document. Integrators can update products via an API in data format. The web search engine receives a message from the management system, and conducts online research, including against competitors, to add to the updated product content information. Based on this research, the system enhances the updated product content. This validation system... The aforementioned management system receives messages from a message unit, which are updated by a vendor or integrator and then enriched with product content in the aforementioned web search engine. Before adding these products to a single product catalog, the system processes their product content by checking for prohibited words and inappropriate images. This is done using the aforementioned vendor product scoring system, which takes into account: the percentage of data entered by the vendor for the relevant product; the number of sales of the relevant product by the vendor; and the rating given to the vendor by customers. The management system takes the vendor's average delivery time data as input and calculates the vendor content score to determine whether the updater (vendor or integrator) has the necessary authorization. Based on the vendor content score calculated by the vendor product scoring system, the management system decides whether the vendor or integrator has the authorization to update. Based on the vendor content score calculated by the vendor product scoring system, the management system either rejects the update request or forwards it to the operation via a service for manual verification. Based on the data processed by the validation system and the update authorization data determined in the relevant step, the management system either approves the update request or forwards it to the operation via a service for manual verification. For requests approved by the management system or the operation, the update data is transmitted to the single product catalog by the management system via a message unit. Our invention specifically involves the product catalog's message unit returning results to the management system, the aforementioned product quality system recalculating the product content quality score for products that have received a positive update request and have been updated in a single product catalog, and the management system displaying/transmitting the request result, along with the newly calculated product content quality score if the update request was approved, or the reason for rejection if the update request was not approved, and any new product content quality score, to the vendor - via a user interface in data format through a service for a single product, in a digital spreadsheet document in data format through a service for a group of products, and to the integrator - via an API in data format through a service. To improve the content quality of individual catalogs used on e-commerce sites, the product quality system uses machine learning to calculate the product content quality score; previous product content quality scores are shown to vendors or integrators; vendors or integrators submit product content update requests to the management system via user interface, documentation, or API; the update content is enriched by a web search engine; the incoming update content is processed in the validation system; the vendor content score is calculated for authorization purposes; the management system decides on update authorization based on the vendor content score; the management system decides to reject or forward the update request to operations; the update request is approved or forwarded to operations for manual verification; for requests approved by the management system or the operations team, the update data is transmitted to the individual product catalog and the result is obtained; the product quality system recalculates the product content quality score and the result of the request is shown/transmitted to the vendor/integrator. This relates to a single catalog product update system that increases productivity. Brief Description of the Figures: Figure 1: A representative visualization modeling the general system structure. Figure 2: A representative visualization of the general system structure, depicting inter-system communication with two-way sequences and modeling algorithmic processes. Reference Numbers 1 Single catalog product update system Product quality system Vendor Integrator 40 Web search robot 50 Validation system 60 Vendor product score system 70 Management system 80 Operation 90 Single product catalog 101 Calculation of product content quality score 102 Displaying a pre-created product content quality score to the vendor or integrator in data format via a service 103 Submitting a product content update request to a management system 104 Enriching the update content 105 Processing the content 106 Calculating the vendor content score 107 Deciding whether update authorization exists 108 Rejecting the update request or forwarding it to operations in data format via a service for manual checking 109 Approving the update request or forwarding it to operations in data format via a service for manual checking 201 For approved requests, sending the update data to the single product catalog via the relevant management system 202 Recalculation of the product content quality score 203 Displaying/transmitting the request result and, if any, the new product content quality score to the vendor/integrator D Document K User interface X All inputs (information) given to the product content system Y All inputs (information) given to the vendor product score system Detailed Description of the Invention Our invention relates to a single catalog product update system and a method (1) related to this, aimed at improving the content quality of products in single catalogs used in e-commerce sites. More specifically; Our invention relates to a system that enables the updating of product information by the same or another vendor (20) in order to maximize the sales of a product whose basic information has been entered into the catalog by a vendor (20), and a method (1) that includes algorithmic processes among the components that make up the said update system. Our invention, the single catalog product update system (1), is basically composed of the product quality system (10), vendor management system (70), operation (80), single product catalog (90) structures and these structures (202,203). API (Application Programming Interface) is defined as a software tool that enables one software to use features defined in another software. REST (Representational State Transfer) is a type of software architecture that emerged in the early 2000s. The REST architecture in question enables fast and practical communication between client and server. This REST architecture allows communication between client and server using data format (XML or JSON). Web services written in accordance with this REST architecture are called RESTful services. Microservices are a type of software architecture where different business tasks are divided into smaller, independent services, and each small service is created to perform only one task. In Figure 2, communication between the vendor (20), integrator (30), operation (80) structures and the management system (70) is provided through a service, and in our invention, a RESTful web service is preferably used. The rest; Product quality system (10), web search robot (40), validation system (50), vendor product score system (60), management system (70) and single product catalog (90) structures are positioned as microservices within the single catalog product update system (1) and perform only a specific function. Communication between these structures is done asynchronously through double-sided queues. Using double-queue data structures between these microservices provides the advantages of scalability, sustainability, testability and flexibility. Here, the asynchronous queue structure is controlled by the message unit, and RabbitMQ is preferably used as the message unit in our invention. This message unit (RabbitMQ) is a message queue system that transmits a message with a specific format (JSON, XML, etc.) from one place to another when it is its turn - at the right time. One of the most important advantages of using the message unit (RabbitMQ) here is its ability to quickly send back a response message when a request message arrives. This results in a very short response time and our system works faster. Also, each cylinder shown in Figure 2 represents a row. In the double-row communication between the mentioned product quality system (10), web search robot (40), validation system (50), vendor product scoring system (60), management system (70) and single product catalog (90), the message unit provides communication and manages the process by sending the messages to the desired location in the correct order. As explained in the previous paragraphs, the vendor (20), integrator (30) and operation (80) structures communicate with each other via a web service. The format of the data required to communicate with the services in question is preferably JSON. Below are examples of JSON request and JSON request result data for the scenario where the mentioned vendor (20) sends an update request to the management system (70) to update the product and the management system (70) responds to the said update request. Example JSON request 2 "ticket/d":"1 1 07835", 3 "merchant":{ "name":"A Store" 7 "products":[ 1 0 "variantSku": "HBC V00002Q5KTP", 1 1 "changes":[ 13 "field":"NAME" 14 "oldVa/ue":"Sol & Luna Leopard Double Side Buckle Set", "newVa/ue":"Sol and Luna Leopard Double Side Buckle Set" 1 8 "field": "DESCRIPTION", 19 "oldVa/ue":"Our product is sold individually. ", "newVa/ue":"Our product is sold as a pair.#diw" Example JSON request result (OCDNOSUI-Äûom`x The scenario mentioned above is more detailed If examined; For the JSON data of the update request; "ticket/d" on the 2nd line represents the update request number and its value is "1107835". "riame" on the . line represents the store name and the store name is "A Store". "variantSku" on the . line represents the product code and the product code is "field" on lines 13 and 18. These sections ("NAME" 9 names, "DESCRIPTION" 9 descriptions) are shown for example purposes and different content sections can also be updated. "oldValue" on the 14th line represents the current name of the product that is to be changed and the current name of the product is "Sol & Luna Leopard Double Side Buckle Set". - Line 15 "newValue" represents the new name of the product that is to be updated, and the new name to be updated is "Sol and Luna Leopard Double Side Buckle Set". - "oldValue" on line 19 represents the current description of the product, and the current description is "Our product is sold as a pair." - "newValue" on line 20 represents the description to be updated, and the description to be updated is "Our product is sold as a pair." - A request is made to the management system (70) for the information in the "NAME" and "DESCRIPTION" fields to be updated by "Store A". For the JSON data of the update request result; - "variantSku" on line 5 represents the product code, and the product code is "HBCV00002Q5KTP". - "qualityScore" on line 6 represents the content quality score of the product, and its value is "8.7". The "field" sections in lines 10 and 17 represent the sections that the seller (20) wishes to update. These sections ("NAME" 9 name, "DESCRIPTION" 9 description) are shown for illustrative purposes only, and different content sections can also be updated. - The "status" sections in lines 11 and 18 represent the approval/rejection status given by the management system (70) for the "NAME" 9 name and "DESCRIPTION" 9 description sections of the product, and the approval status for both sections is "APPROVE" 7. - The "oldValue" name in line 12 represents the current name of the product that is to be changed, and the current name of the product is "Sol & Luna Leopard Double Side Buckle Set". - The "newValue" name in line 13 represents the updating of the product that is to be changed. - The desired new name is represented by "oldValue" on line 19, and the new name to be updated is "Sol and Luna Leopard Double Side Buckle Set". - The current description of the product is represented by "oldValue" on line 19, and the current description is "NewValue" on line 20, and the description to be updated is "Our product is sold as a pair." More specifically, the management system (70) reports the product with code "HBCV00002Q5KTP" as "NAME" and separate "status". The approval/rejection status is included by the management system (70). From here on, the operation of the system and the interaction of the system parts with each other will be explained. The mentioned product quality system (10) has fixed values for each field such as product name, number of images, description content, etc. for different product types for the products already in the single product catalog (90). The product has a pre-established product content quality score based on the scores determined by the decisions. In order to change this score, the product quality system (10) calculates a product content quality score. This calculation input is made for newly made or updated products. The above-mentioned calculation is done using machine learning methods to determine which information contributes to sales for each category. Here, the product quality system (10) is a machine learning model. The information given below - product name - number and quality of visuals - presence of video - length of product description - whether the product description contains visuals, videos, tables and lists - whether product type-based feature values (color, size, etc.) are entered are given as input to the machine learning model and the model is fed with the data on how many units were sold in return. The information given to and fed to the machine learning model is: All inputs are represented as (X) in Figure 2. Furthermore, the machine learning model calculates the product content quality score only once for each product in the catalog. Additionally, the model calculates the product content quality score for the first time for products added to the catalog and again for updated products. More specifically, the product quality system (10) takes the following data as input: product name; number and quality of visuals; presence or absence of video; length of product description; presence or absence of visuals, videos, tables, and lists in the product description; entry or non-entry of product type-based feature values (color, size, etc.); then it takes the data on how many sales were made according to this data as input; and by taking the two different data(s) mentioned in the previous two sentences as input, it calculates the product content quality score once for each category in the single product catalog (90) in order to determine which data contributed to product sales. It calculates (101), transmits the calculated score as an output to the management system (70) via the message unit (RabbitMQ), and recalculates the content quality score for products that have a positive update request and are updated in the single product catalog (90) (202). The mentioned product quality system (10) shows the product content quality score, which is created in advance based on fixed scores for each field such as product name, number of images, description content, etc., for different product types in the mentioned single product catalog (90), to the vendor (20) or integrator (30) via a service (RESTfuI) in data format (JSON) (102), after which the vendor (20) or integrators (30) can update the product data and increase the product content quality score. When the said vendor (20) wants to make an update for a single product, the user interface (K) When the mentioned integrators (30) want to make a bulk product update, they send the update request along with the update information to the management system (70) using the spreadsheet document (D). When the mentioned integrators (30) want to make a product update, they send the update request and update information to the management system (70) using the APIs (A) opened by the management system (70). More specifically; when the mentioned vendor (20) wants to make an update for a single product, they send the update request to the management system (70) via a service (RESTfuI) in the form of a user interface (K) in the form of data format (JSON), when the mentioned vendor (20) wants to make a bulk product update, they send the update request to the management system (70) via a service (RESTfuI) in the form of a spreadsheet document (D) in the form of data format (JSON), and when the mentioned integrator (30) wants to make a product update, they send the product content update request to the management system (70) via a service (RESTfuI) in the form of data format (JSON) using the API (A). It transmits (103). The product data updated by the mentioned vendor (20) or integrators (30) is transmitted from the management system (70) to the web search robot (40) via the message unit (RabbitMQ). Then, the web search robot (40) conducts research on the internet, including competitors, enriches the product content by adding to the product content information updated by the vendor (20) or integrator (30) according to the results of this research (104), and transmits the enriched product content to a management system (70) via the message unit (RabbitMQ). Here, the web search robot (40) is an internet content search robot (google, yandex, etc.) that brings together and reports the data found on the internet. More specifically, the web search robot (40) receives the data from the aforementioned management system (70) via the message unit (RabbitMQ). A vendor (20) or an integrator (30) searches the internet, including competitors, to add to the product content information that has been updated, and enriches the updated product content (104) by adding to the product content information based on the results of this research, and transmits the enriched product content to a management system (70) via a message unit (RabbitMQ). In the aforementioned web search robot (40), the enriched updated content was transmitted to the management system (70). The management system (70) transmits the enriched updated content to a validation system (50) via a message unit. The enriched updated content is processed in the aforementioned validation system (50) (105). Thanks to the aforementioned validation system (50), it is ensured that the products whose product data has been updated by the aforementioned vendor (20) are not immediately reflected in the catalog. Thus The products updated by the seller (20) are not immediately reflected in the catalog. Similarly, the update data entered by the integrator (30) is checked by the validation system (50), and the updated products are not immediately reflected in the catalog. This validation system (50) provides checks such as word checks, inappropriate visual checks, etc., and produces a result based on these checks. In this way, brand owner and single seller products are updated in the catalog as a result of the basic validation process. The aforementioned validation system (50) processes the content of the products that it receives from the management system (70) via the message unit (RabbitMQ) and that are updated by a seller (20) or an integrator (30) and then enriched in the web search robot (40), by checking their content with word checks and inappropriate visual checks before adding them to the single product catalog (90) (105). More specifically, the validation system (50) checks the updated product content – which is received from the management system (70) via a message unit (RabbitMQ) and updated by a vendor (20) or an integrator (30) and then enriched in the web search engine (40) – before the updated product content is added to the aforementioned single product catalog (90) – through checks (forbidden word check, inappropriate image check, etc.) and transmits the result it produces to a management system (70) via a message unit (RabbitMQ). The vendor product score system (60) has a vendor content score that is created based on fixed scores determined by decisions for each area, such as the rate of data entered by the vendor (20) for the relevant product, the number of sales of the relevant product by the vendor (20), etc. Here, the aforementioned In order to change the score, the seller product score system (60) calculates the seller content score (106). Here, the seller content score is calculated for sellers who have recently joined or whose information has been updated. This calculation is done using a machine learning method to determine whether the updater has the authority. Here, the seller product score system (60) is a machine learning model. The information specified below (Y), - the rate of data entered by the seller (20) for the relevant product - the number of sales of the relevant product by the seller (20) - the rating given to the seller (20) by customers - the average delivery time of the seller (20) are given as input to the machine learning model and the seller content score is calculated in return (106). More specifically, the seller product score system (60) calculates the seller content score based on the rate of data entered by the seller (20) for the relevant product; the number of sales of the relevant product by the seller (20). The system calculates the seller content score (106) to determine whether the updater (seller (20) or integrator (30)) has the authority to update, taking into account the sales quantity, customer ratings given to the seller (20), and the average delivery time of the seller (20) as data inputs, and transmits the calculated score as an output to the management system (70) via the message unit (RabbitMQ). All four data inputs mentioned above are represented as input (Y) in Figure 2. The management system (70), taking into account the seller content score calculated by the seller product score system (60) described in the previous paragraph, decides whether the seller (20) or the integrator (30) has the authority to update (107). Then, the management system (70), taking into account the seller content score calculated by the seller product score system (60) described in the previous paragraph, decides whether the seller (20) or the integrator (30) has the authority to update (107). By receiving it, the management system (70) rejects the update request made by the vendor (20) or integrator (30) or directs it to the operation (80) in data format (JSON) via a service (RESTfuI) for manual checking (108). The management system (70) approves the update request by taking into account the data processed by the aforementioned validation system (50) and whether there is update authorization decided in step -107- or not, or directs it to the operation (80) in data format (JSON) via a service (RESTfuI) for manual checking (109). For requests approved by the management system (70) or the operation (80) team, the update data is transmitted to the single product catalog (90) by the management system (70) using the message unit (RabbitMQ) and the single product catalog (90) is transmitted to the management system via the message unit (RabbitMQ). (70) The result is returned (201). Afterwards, the approved products are updated in the single product catalog (90). The single product catalog (90) mentioned here is a system where the saleable version of all products is kept with all category and product information. Subsequently, the product quality system (10) recalculates the product content quality score for the products that have received a positive update request and have been updated in a single product catalog (90) (202). The product content quality score calculated for the second time includes the processes used when calculating the first product content quality score. Unlike the first calculation, it is expected that the product content quality score will increase positively because an update has been made. Finally, the request result and, if any, the new product content quality score are shown/communicated to the vendor/integrator (203). Here, the management system (70) sends the request result and, if any, the new product content quality score to the vendor (20) - single product catalog (70); if the update request is approved, the newly calculated product content quality score, if the update request is not approved, the reason for rejection, and, if any, the new product content quality score to the vendor (20). In its most general form, the management system (70) decides whether to reject the update request made by the vendor (20) or integrator (30), to send it to the operation team (80) in the form of data format (JSON) via a service (RESTfuI) and, if positive, to send the updated product to the single product catalog (90) via the message unit (RabbitMQ), taking into account the vendor content score calculated by the vendor product scoring system (60), to send it to the operations team (80) in the form of data format (JSON) via a service (RESTfuI) for manual checking, or, if positive, to send the updated product to the single product catalog (90) via the message unit (RabbitMQ). In addition, the management system (70) (70) If the update request is approved, the management system (70) communicates/shows the request result, along with the newly calculated product content quality score, and if applicable, the new product content quality score, to the vendor (20) via the screen interface. If the update request is not approved, the management system (70) communicates/shows the request result, along with the reason for rejection, and if applicable, the new product content quality score, to the integrator (30) via API (A). Here, the integrator (30) can obtain information such as the request result, product content quality score, and reason for rejection by querying at certain intervals using API (A). After all the processes mentioned above, the product content quality in the single catalog is improved.

Claims (1)

STEMLER 1. The invention is a single catalog product update system (1) to be used in single catalogs within e-commerce sites, and its feature is; in order to increase the quality of product content in single catalogs and to maximize product sales accordingly; product name; number and quality of images; whether there is a video; length of product description; whether the product description contains images, videos, tables and lists; At least one product quality system (10) that takes the data of product type-based feature values (color, size, etc.) as input, then takes the data of how many sales were made according to the data in question as input, calculates the product content quality score once for each category in a single product catalogue (90) in order to determine which data contributes to the product sales by taking the two different data/data mentioned in the previous two sentences, transmits the calculated score as an output to the management system (70) via the message unit and re-calculates the content quality score for the products whose update request is positive and updated in a single product catalogue (90), conducts a research including competitors on the internet in order to add to the product content information updated by a seller (20) or an integrator (30) by using the message unit from the said management system (70), and conducts the said research at least one web search robot (40) that enriches the product content by making additions to the product content information in question according to its results and transmits the enriched product content to a management system (70) via a message unit; before the products with updated product content are added to a single product catalogue (90), the said update content - product content that is updated by a seller (20) or an integrator (30) and whose content is later enriched in the said web search robot (40) - undergoes checks (forbidden word check, inappropriate visual check, etc.) and transmits the result it produces following the said check to a management system (70) via a message unit; at least one validation system (50) that includes the data rate entered by the seller (20) for the relevant product; the sales quantity of the relevant product by the seller (20); The score given to the vendor (20) by the customers is characterized by the components of at least one vendor product score system (60) that takes the average delivery time data of the vendor (20) as data input, calculates the vendor content score in order to decide whether the person making the update (vendor (20) or integrator (30)) has the authority with the data inputs it receives as 0 and sends the calculated score as an output to the management system (70) via the message unit; at least one management system (70) that decides whether the update request made by the vendor (20) or integrator (30) is rejected, forwarded to an operation (80) team in data format via a service for manual control or, if positive, the updated product information is transmitted to a single product catalogue (90) via the message unit. According to claim 1, there is a single catalog product update system (1), and its feature is; said product quality system (10) is a machine learning model. According to claim 1, there is a single catalog product update system (1), and its feature is; said web search robot (40) is an internet search robot that brings together and reports the data found on the internet. According to claim 1, there is a single catalog product update system (1), and its feature is; said seller score system (60) is a machine learning model. According to claim 1, there is a single catalog product update system (1), and its feature is; said message unit is RabbitMQ. According to claim 1, there is a single catalog product update system (1), and its feature is; the data format used to communicate with said services is JSON. The invention is a single catalog product update method (1) to be used in single catalogs within e-commerce sites, and its feature is; In order to increase the product content quality in single catalogues and to maximize product sales accordingly; the mentioned product quality system (10) takes as input the data of -product name; number and quality of visuals; presence of video; length of product description; presence of visual, video, table and list in product description; product type based feature values (color, size, etc.) - and subsequently the data of how many sales were made according to the mentioned data, and in order to determine which data contributes to product sales, it calculates the product content quality score (101) once for each category in a single product catalogue (90), the said product quality system (10); for the products currently in the said single product catalogue (90), it calculates the product content quality score (101) in different product types in terms of product name, number of visuals, description content, etc. showing a product content quality score (102) previously created based on the scores determined with fixed decisions for each field to the vendor (20) or integrator (30) in data format through a service; when the said vendors (20) want to make an update for a single product, they send it in data format via a user interface (K) through a service; when they want to make a mass product update, they send it in data format via a digital table document (D) through a service; when the said integrators (30) want to make a product update, they send the product content update request to a management system (70) via API (A) through a service (103); the said web search robot (40); in order to add to the product content information updated by a seller (20) or an integrator (30) via the message unit from the said management system (70), it conducts research on the internet including competitors and makes additions to the said product content information according to the results of the said research, thus enriching the update content (product content) (104); the said validation system (50); processes the content of the products received via the message unit from the said management system (70) and updated by a seller (20) or an integrator (30) and later enriched in the said web search robot (40) by putting the product contents of the said products through the following checks; o forbidden word check o inappropriate visual check before being added to a single product catalogue (90); the said seller product scoring system (60), - the data rate entered by the seller (20) for the relevant product; the seller (20) takes the relevant product sales number; the score given to the seller (20) by the customers; the average shipping time of the seller (20) as data input and calculates the seller content score (106) in order to decide whether the person making the update (seller (20) or integrator (30)) has the authority; the management system (70) decides whether the seller (20) or integrator (30) has the authority to update, considering the seller content score calculated by the said seller product score system (60) (107); the management system (70) rejects the update request made by the seller (20) or integrator (30) or directs it to the operation (80) in the form of a data format through a service to be checked manually (108). said management system (70), taking into account the data processed by said validation system (50) and whether it has the update authorization decided in step -107-, approves the update request or directs it to the operation (80) in data format via a service to be checked manually (109). for requests approved by said management system (70) or operation (80), transmission of update data to the single product catalogue (90) by said management system (70) via the message unit and return of result to the management system (70) via the message unit of said single product catalogue (90), recalculation (202) of product content quality score by said product quality system (10) for products whose update request is positive and updated in a single product catalogue (90) and said management system (70); If the update request is approved, it is characterized by displaying/transmitting the request result and, if any, the new product content quality score together with the newly calculated product content quality score, if the update request is not approved, together with the reason for rejection, to the vendor (20) 0 for a single product via a service in the form of a data format via a user interface o for a group of aggregated products via a service in the form of a digital table document (D) to the integrator (30) o via a service in the form of a data format via API (A) (203).